Summary and Conclusions

Water-resources management in the Everglades has evolved from the original purpose to protect against floods and to conserve water during dry periods, to also include protecting and restoring natural functions and characteristics. The main effect that digging canals and draining wetlands had was to reduce the water-storage capacity of this subtropical peatland. Over time, the decreased water storage contributed to the degradation of a natural balance between highly seasonal water flows, pristine low-nutrient freshwaters, and a unique assemblage of specially adapted plants and wildlife. The first half of the twentieth century saw the many adjustments to the changing hydrology that caused a general drying out of the Everglades system. The second half of the twentieth century began with the completion of a system of levees that enclosed Water Conservation Areas (WCAs). For a number of years following completion of WCAs, water tended to be stored at levels that were too high for the ecosystem during wet periods, resulting in problems such as the drowning of tree islands. Equally undesirable was an approach to managing floods that involved conveying vast amounts of freshwater (that the pre-drainage Everglades previously could have stored) through canals to be flushed to the Atlantic Ocean. During excessively dry periods control structures were shut to keep water levels at minimum acceptable levels in WCAs, which had the effect of curtailing the southerly flow of water to the Everglades National Park (ENP). Even during times between storms and droughts, problems persisted with water management in the Everglades. Some of these problems included shallow seepage beneath levees that occurred at all times and deeper ground-water flows increased from the Everglades to outside areas after an extended period of ground-water pumping in well fields to the east. Thus, southerly flows to ENP were depleted even during times of typical water-level conditions. The Comprehensive Everglades Restoration Plan, authorized by Congress in 2000, sought to address the large, complex, and expensive to remediate water-management problems.

Except for seepage studies beneath the eastern boundary levees, interactions between surface water and ground water have not been extensively investigated in the Everglades. More work is needed on accurately estimating recharge and discharge, and the effects on the water balance and water quality of the Everglades. Investigations of the factors that control recharge and discharge also are important. A problem of special concern in the Everglades is characterizing recharge and discharge in the vast wetland interior areas, where few studies have been completed. Interactions between surface water and ground water in interior areas, and their modification because of water-resources management, are relevant to the protection of fresh ground-water resources beneath the vast interior of the Everglades.

In an effort to learn more about surface-water and ground-water interactions in the Everglades, the U.S. Geological Survey (USGS) and South Florida Water Management District (SFWMD) developed an agreement to undertake a detailed study of those interactions in selected areas of the north-central Everglades. The present study focused its efforts on the most highly manipulated part of the Everglades system–areas of the north-central Everglades that had been converted to farming during the past century and now returned to management as constructed wetlands, called Stormwater Treatment Areas (STAs).

The present investigation determined the extent to which interactions between surface water and ground water have increased in the north-central Everglades as a result of water-resources management. On the western side of the WCAs, and in the Stormwater Treatment Areas, or STAs, there has been a dramatic shift since pre-drainage times in the direction of horizontal ground-water flow and in the magnitude of recharge. The direction of ground-water flow shifted southwest to northwest as a consequence of the steepening of land and water-table slopes that followed extensive land subsidence in the Everglades Agricultural Area, or EAA. Another factor that was important in changing interactions between surface water and ground water was the abrupt change in water level (typically 2.5 ft or more) across levees that increased driving forces for recharge and discharge. Rapid releases of large quantities of surface water between basins also had the effect of increasing recharge and discharge by propagating surface waves (and subsurface pressure waves through the aquifer), causing temporary periods of increased recharge often followed by temporary periods of increased discharge.

Recharge is the dominant interaction between surface water and ground water in the northern Everglades when averaged over space and time. Net recharge is the result of increased hydraulic gradients caused by land subsidence and decreasing water-table elevations outside of the WCAs. Basin-averaged recharge at the Everglades Nutrient Removal (ENR) project (0.9 cm/d) exceeded recharge at Water Conservation Area 2A (WCA-2A) (0.2 cm/d) by more than a factor of 4. Recharge exceeded discharge in both ENR and WCA-2A, by a factor of 10 at ENR, and by a factor of 2 in WCA-2A. Recharge in both ENR and WCA-2A accounted for a volumetric flux of water equal to approximately 30 percent of surface water pumped into each basin.

Recharge and discharge increased since pre-drainage conditions in the vicinity of ENR as a result of water-management conditions. The estimated change was from approximately no net flux under pre-drainage conditions to 0.9 cm/d net recharge under present water-management conditions. Changes in net recharge and discharge probably were not as dramatic in WCA-2A, from no net flux under pre-drainage conditions to approximately 0.08 cm/d net recharge under present conditions. An increase in net recharge since pre-drainage conditions mainly has been controlled by large-scale and long-term changes, such as subsidence in the agricultural areas and increasing ground-water pumping from well fields to the east of the Everglades.

Site-specific estimates of recharge and discharge in ENR and WCA-2A ranged between detection limits (0.04 cm/d) and 20 cm/d. The general pattern of recharge and discharge measurements showed that interactions between surface water and ground water decrease from the north Everglades toward the north-central Everglades. The largest recharge and discharge fluxes were observed in ENR. From ENR, recharge and discharge decreased consistently toward the wetland interior of WCA-3A. What explains the pattern of decrease from north to south? Recharge and discharge fluxes were highest within 0.5 km of levees (always greater than 0.3 cm/d), whereas vertical fluxes in the wetland interior rarely exceeded 0.3 cm/d. The order of magnitude difference in vertical fluxes near and far from levees suggests that basin-averaged fluxes are predictable from the ratio of the length of levee perimeter to the surface area of each basin, which decreases from the northern extent of the Everglades (ENR and STAs) towards the WCA-3. Greater hydraulic conductivity in WCA-2A peat compared with ENR also affected recharge and discharge. Differences in hydraulic properties of peat appear to be explained by peat compaction in wetlands managed for agriculture for appreciable time (decades).

Vertical fluxes in the interior areas of the wetland basins were relatively small and varied over time, frequently changing direction. To put these fluxes in the perspective of other water balance fluxes, recharge and discharge in interior areas were usually less than 50 percent of average precipitation and evapotranspiration (0.4 cm/d and 0.35 cm/d, respectively). Nevertheless, recharge and discharge in the wetland interior still are quite important. For example, recharge in the interior of WCA-2A accounted for a volumetric flux equal to 30 percent of surface flow through that basin.

Little is known about the factors that control vertical exchange in the wetland interior. Hydrogeologic model simulation showed that recharge and discharge in the wetland interior are not explained by water-level differences across levees. Instead, it is more likely that the local and regional water-surface slopes, controlled by topography and water releases between basins, have a greater effect on recharge and discharge in the wetland interior. The factor discussed in great detail in this report was the effect of large releases of water in WCA-2A that transmit waves of surface water through the wetlands, and induce pressure gradients that cause cycles of recharge and discharge. Another important factor is the layering of different aquifer sediments in the aquifer, which transmits surface-pressure perturbations at different rates. The relation between water releases and hydraulic properties of aquifer sediments and peat drives vertical exchange in ways that are not yet fully understood. It is clear, however, that recharge and discharge are related closely with fluctuating surface-water levels. Reversal between recharge and discharge in the wetland interior may not affect the long-term average water balance; however, those vertical fluxes could contribute to net exchange of solutes, including contaminants, between wetlands and ground water. Future investigations will require better field measurements and more sophisticated modeling to test the importance of such factors in controlling surface-water and ground-water interactions.

Geochemical evidence provides a long-term integrated picture of hydraulic processes in the aquifer and exchange with surface water. Water-stable isotopes indicated that the major source of fresh water in Everglades ground water is recharge of evaporated surface water from the wetlands rather than infiltration of rainfall through unsaturated sediments. This observation is consistent with the interpretation that both recharge and discharge have always occurred in the Everglades, with recharge occurring at time periods of relatively high surface-water levels and discharge occurring when surface-water levels were low.

Major ion chemistry and water-stable isotopes identified eight major water types in the north-central Everglades. Four of those types were considered to be source waters. The other four are mixtures of the source waters. The distribution of those water types identified long-term average flow pathways of recharge and discharge. Particularly evident as a major factor driving recharge and discharge in the present-day Everglades was the effect of wetland compartmentalization by levees. Geochemical water typing provided information that was consistent with hydraulic information. The geochemical tracers, however, particularly were useful in showing the extent to which water management has caused vertical mixing throughout essentially the entire depth of the Surficial aquifer. The effect of levees is indicated by the presence of Fresh Recharge Water recharged to a depth of at least 90 ft on the up-gradient (headwater) side of the levee, and Relict Seawater discharged on the down-gradient (tailwater) side of the levee. The effect of levees on interactions between surface and ground water, therefore, extends vertically throughout at least the top half of the Surficial aquifer (100 ft), and horizontally for at least 1 mi away from the levees. These observations support the hydraulic interpretation that water management has increased interactions between surface water and ground water in the north-central Everglades.

A wide-ranging assessment of the distribution of mercury in surface water, peat porewater, and ground water found no evidence that ground water was a significant source of mercury to surface water. Rather, peat and surface water appear to be a source of mercury to ground water, transported with recharging ground water. Maximum values of total mercury (Hg_T) in shallow ground water were approximately equal to corresponding measurements in surface water at both ENR and WCA-2A. Methylmercury (MeHg) was detectable only in certain shallow wells at ENR and WCA-2A. Those wells had MeHg concentrations that were 50 percent and 10 percent of the highest MeHg concentrations in surface water at ENR and WCA-2A, respectively. Whereas Hg_T was detectable throughout the Surficial aquifer, MeHg was mainly present at detectable concentrations only in ground water recharged at the ENR site when that area was managed for agriculture.

Variable SO₄ concentrations do not co-vary with MeHg concentrations at either ENR or WCA-2A. Source of recharge water generally is thought to cause variability in mercury concentrations. For example, MeHg only was detectable in shallow ground water beneath ENR in wells where major ion chemistry and water-stable isotopic ratios indicated that agricultural recharge water was the source. Eight of 17 samples from wells with agricultural recharge water had detectable MeHg, with values as high as 0.2 ng/L. In contrast, well samples with wetland recharge water all had very low values of MeHg (less than 0.04 ng/L). Type of recharge water was not as important for Hg_T, because measurements greater than 0.5 ng/L were distributed equally between the agricultural recharge water and wetland recharge water.

Ground-water fluxes of mercury were determined as part of a steady-state hydrologic and mercury balance in ENR. The resulting mercury flux computations indicated that ground-water discharge to ENR from the eastern side (WCA-1) was not appreciable, contributing approximately 0.4 and 0.3 percent of Hg_T and MeHg to ENR, respectively. However, recharge on the western side of the ENR was a major pathway for export of Hg_T from ENR. Recharge of Hg_T accounted for an export of 68 g/yr, or 10 percent of all inputs to ENR. Recharge of MeHg was negligible, either below detection, or, if values of MeHg close to detection levels are used, accounting for no more than export of 1 g/yr (3 percent of inputs).

Based on the mass budget for Hg and other evidence, it was concluded that recharge of Hg_T was augmented with release of more Hg_T during downward transport through the peat. The primary evidence for this conclusion is that average concentrations of HgT measured in shallow ground water were higher than average values measured in surface water. Independently measured profiles of Hg_T in peat porewater corroborate that conclusion, showing that Hg_T concentrations are high throughout the entire depth of peat. Therefore, it appears that solid-phase mercury stored in the peat is being mobilized and recharged to the aquifer. Mass-balance estimates suggest that 70 percent of the recharged Hg_T comes from surface water, whereas 30 percent comes from release from the peat. In contrast to results for Hg_T, it was found that MeHg was not recharged to the aquifer. Support for this conclusion comes from measurements of MeHg in shallow ground water that are below detection, and from porewater profiles in peat that show relatively high concentrations of MeHg are restricted to very near the peat surface.

Chemical data and water-stable isotopic ratios indicate that most surface water recharged in ENR is discharged to a seepage canal on the western and northern side of ENR. Transport of recharged water through the aquifer to the seepage canal appears to take place in a matter of weeks to months, with only relatively minor mixing with deeper ground water. Measurements of Hg_T in the seepage canal suggested that Hg_T had not yet discharged to the canal at the end of the 4-yr study period (1994-98). Because the travel time between points of recharge in ENR and discharge in the seepage canal was relatively short (weeks to months), it was concluded that mercury was retained or delayed in its transport through the aquifer by interaction with aquifer sand or limestone or fine organic materials at the base of the seepage canal.